proteins of interest
To demonstrate the capabilities of the degron approach and to further characterize the degron’s behavior on a molecular level, different reporters were fused to the optimized N-degron cassette (K2, Gowda et al. 2013, Fadenet al. 2016b). All reporter proteins offered an easy readout and/or were well established proteins that have been used widely in a bio-logical/biochemical context, therefore being of general interest to the science community1. To confirm true regulation on protein level, transcripts of many of the reporter lines were analyzed.
3.1.1 Generation of a K2:GUS expressing reporter line
β-glucuronidase (GUS) is an enzyme fromE. coli that has been used extensively in higher plants for promoter and expression studies (Jeffersonet al., 1987). It catalyzes the hydrol-ysis of β-glycosidic bonds, which in enzyme assays with either 5-bromo-4-chloro-3-indolyl glucuronide (X-Gluc) or 4-methylumbelliferyl-beta-D-glucuronide (4-MUG) results in eas-ily measurable read-outs which are a bright blue precipitate for X-Gluc and a fluorescence signal for 4-MUG. K2:GUS expression in stably transformedA. thaliana plants was driven by either ProCaMV35S (Pro35S), ProUBQ10, or ProCDKA;1.
Identification of responsive K2:GUS expressing lines: T2 plants were screened for degron functionality through a qualitative GUS assay of plants grown at 14°C or 28°C.
In the initial screen, two lines, one for the CDKA;1 and one for the UBQ10 promoter construct, were identified that showed a temperature dependent accumulation of the active GUS protein, as made visible by a blue color precipitate. However, expression of K2:GUS in the ProCDKA;1 controlled line seemed weak (fig. S 5.3B). Most of the other lines either did not show any color, or showed color independent of the growth temperature. Mainly Pro35S driven lines behaved in this way, most likely due to strong expression from the viral promoter (fig. 3.1A). The two identified lines were grown again at 14°C or 28°C and
1An overview over this first generation of K2-reporter constructs, depicting overall organization as well as used promoters and primary destabilizing residues, can be found in fig. S 5.2.
subjected to a quantitative GUS assay. The ProCDKA;1 line proofed to be a false positive since it did not show significant GUS activity (fig. S 5.3C).
Protein analysis: To evaluate whether the qualitative differences would correlate with abundance of GUS protein, samples were generated from cold and warm grown plants and subjected to Western Blot analysis. Signals for the K2:GUS fusion protein always appeared specifically as two distinct species over and below 100 kDa (fig. 3.2B). The lower molecular weight signal occured at the in silico predicted mass weight of the K2:GUS fusion protein of 94.3 kDa. Longer exposure of blots always resulted in both signals being visible in both samples (14°C or 28°C) with different intensities (fig. S 5.3D). The higher molecular weight signal was found to be stronger in samples extracted from cold grown plants, whereas the lower molecular weight signal was usually more abundant in samples isolated from warm grown plants. Longer exposure led to both signal appearing in both samples, whereas weaker exposure of the blot led to a characteristic ”zig-zag” scheme of bands (figs. 3.1B and 3.2).
3.1.1.1 Characterization of the ProUBQ10:K2:GUS reporter line
A qualitative GUS assay confirms the response of K2:GUS levels to tem-perature: To analyze the GUS activity, a Pro35S:GUS2, as well as a non-responsive Pro35S:K2:GUS expressing line were used as controls. GUS activity in the two control lines was significantly higher then in the degron expressing lines. Also the dynamics of the GUS accumulation was inversed. Were non-responsive lines accumulated more active GUS enzyme under warm conditions, the ProUBQ:K2:GUS expressing line accumulated significantly more active GUS enzyme when grown at 14°C then when grown at 28°C, in-dicating functionality of the degron in regard to temperature dependent accumulation of the K2:GUS fusion protein (fig. 3.1A).
A quantitative GUS assay reveals inverse accumulation kinetics between the responsive ProUBQ10:K2:GUS expressing line and non-responsive control lines: Plant lines used for the qualitative assay were now subjected to quantitative analy-sis. GUS activity in the two control lines was found to be 3 to 23 times higher then in the K2:GUS expressing lines. Also, the dynamics of the GUS accumulation was inverted. Were non-responsive lines accumulated more active GUS enzyme at 28°C, ProUBQ10:K2:GUS reliably accumulated significantly more active GUS enzyme at 14°C (fig. 3.1C). Also it appeared as if the fusion of the degron to the GUS enzyme influenced overall protein levels as the activity in the Pro35S:K2:GUS expressing plants was significantly lower compared to the GUS activity found in the Pro35S:GUS expressing lines. This data confirms the results obtained in the qualitative GUS assay (fig. 3.1A)
Temperature shift experiments show a quick and timely responsive of K2:GUS
2Expression from the binary plasmidpBI121 (Jeffersonet al., 1987); kind gift of Diana Schmidt (Leibniz Institute of Plant Biochemistry, Halle)
Figure 3.1 – The degron cassette stirs K2:GUS accumulation in a highly re-sponsive, temperature dependent manner.(A)Qualitative GUS assays of different lines expressing K2:GUS under the control of different promoters reveal a temperature dependent stability only when expression is driven by the weaker UBQ10 promoter. (B) Western Blot analysis of cold and warm grown, K2:GUS expressing, plants reveal double banding with a typical zig-zag signal appearance. Blots were probed with anti-Ha (tar-geting the degron cassette) and anti-GUS (directly tar(tar-geting the GUS moiety) antibodies respectively. (C)Quantitative analysis confirms the observation made in (A). N = 6 for ProUBQ:K2:GUS and N = 3 for all others. p<0.5 *, p<0.01 **. p<0.001 ***. unpaired ttest. (D)The degron reacts to a temperature shift within hours. Stabilization at 14°C is not completed 24 hours past shift. N = 3(E)Degron stability and linked GUS activity is flexible and tunable following a temperature stimulus in teh transgenic GUS-expressing plants. N = 3
levels to altered temperatures: To dissect the resolution of the degron accumulation in regard to temperature and time, two different shift experiments were employed. In a first approach, plants were grown at 21°C temperature on plates. After two weeks, plates were shifted to either 14°C or 28°C respectively. Upon temperature shift, activity increased or decreased in a temperature dependent manner, indicating accumulation or degradation of the K2:GUS fusion protein (fig. 3.1D). The level of GUS activity dropped quickly and reached a low steady-state already after eight hours as indicated by reached statistically significant differences in activity as compared to time point zero (for values see additional data in table 5.1). The changed activity correlated with the total amount of K2:GUS fusion protein (fig. S 5.3D).
A temperature tuning experiments shows a linear link between K2:GUS sta-bility and temperature: Additionally to the first experiment, where the reaction time of the stabilization of the K2:GUS fusion protein was assessed, a second experiment was performed to show tuneability as well as (de)stabilization kinetics of the K2:GUS fusion protein. Plants were grown at 14°C or 28°C. Every 24 h a sample was taken and the temperature shifted for 2.8°C up or down. The degron quickly responded to changed tem-peratures, which allows to tune protein content to desired levels (fig. 3.1E). The response of the degron to switching temperatures seemed linear, thus confirming the results of the temperature shift experiment (fig. 3.1D/E).
An inhibitor treatment indicates proteasomal degradation of the K2:GUS fusion protein: The previous experiments indicated that K2:GUS stability, and there-fore the enzymatic GUS activity per µg of total protein, is linked to temperature. How-ever, it remained unclear whether the K2:GUS fusion protein is readily degraded by the Ubiquitin Proteasome System (UPS) as predicted. Treatment of seedlings with the pro-teasomal inhibitor MG132 clearly showed accumulation of higher molecular weight signals after treatment of plants grown at 28°C. This indicates that the K2:GUS fusion protein is indeed degraded by the proteasome. Also, it suggests that the higher molecular species of K2:GUS represent the fraction of total K2:GUS protein, that confers the activity, sensitive to temperature dependent accumulation/degradation (fig. 3.2A).
Figure 3.2– Inhibitor treatments, enrichment, and mass spectrometric anal-ysis suggest potential distinct ubiquitinated species of K2:GUS and its degradation via the UPS.(A) Treatment of stable transgenic seedlings expressing ProUBQ10:K2:GUS with the proteasomal inhibitor MG132 indicates degradation of the K2:GUS fusion protein via the proteasome as indicated by appearance of higher molecu-lar species after treatment (white triangles). (B)Immunoprecipitation of K2:GUS form stable transgenicProUBQ10:K2:GUS expressing plants using a DHFR specific antibody.
The left panel shows success of the pulldown. Not only the two previously identified signals appear but also numerous others of intermediate molecular masses indicating a higher diversity of different sub-species. The right panel indicates the three different samples (high, middle, low) cut from the gel and subjected to mass spectrometric analy-sis. Peptides specific for the fusion K2:GUS fusion protein were identified in all samples (compare Faden et al. 2016b for exact results). (C) Ubiquitome enrichment of sta-ble transgenic plants expressing ProUBQ10:K2:GUS, using a TUBE matrix, indicates that the multiple signals observed in K2:GUS expressing plants indeed represent various ubiquitinated subspecies of the fusion protein. Especially the higher molecular weight species are enriched in the elution fraction when the starting material was grown at cold temperatures.
Immunoprecipitation (IP) and Mass Spectrometry confirm the identity of the different protein sub-species: To analyze the different species of K2:GUS, the fusion protein was immunoprecipitaded and subjected to mass spectrometric analysis. The success of the IP was monitored via SDS-PAGE. For better resolution of the high molecular weight signals, a low acrylamide percentage of 8% was chosen. The number of K2:GUS species, as indicated by the different specific signals appearing, seemed to be even higher than initially observed. Whereas other Western Blots always showed a separation into only two species (figs. 3.1A, 5.3D, and 3.2A/C), the analysis on a low percentage gel suggested at least four species of the K2:GUS fusion protein (fig. 3.2B left panel). However, the two species previously identified were the most abundant ones. To test whether these species contained the K2:GUS fusion protein the larger part of the IP reaction was loaded on another gel and silver-stained. Three different parts of the gel were cut out and subjected to mass spectrometric analysis (fig. 3.2B left panel). Indeed, in all three fractions peptides specific for the GUS enzyme could be identified (compare Faden et al. 2016b).
Enrichment of the Ubiquitome of K2:GUS expressing plants reveal over-representation of discrete sub-populations of the ubiquitinated fusion protein:
To elucidate the ubiquitination state of the K2:GUS fusion protein under different tem-peratures, a Tandem Ubiquitin Binding Entities (TUBE) based enrichment of the entire ubiquitome of plants grown at 14°C or 28°C was carried out. Wild type plants were used as a control to ensure specificity of the immunosignals in the elution fraction. Especially the higher molecular weight signals always appeared specifically enriched in the TUBE-enriched ubiquitome fraction (fig. 3.2C). Additionally, a typical ubiquitination smear is visible atop the upper K2:GUS elution signal. This might indicate that the higher band of the two bands always visible in the K2:GUS line represents a specific state of ubiqui-tination. Interestingly, the control blot probed with an anti-Ub antibody revealed that the overall abundance of ubiquitinated proteins seemed to be lower in plants grown at 28°.
Also no difference between transgenic and wild type plants in regard to total ubiquitination levels was apparent.
3.1.2 Cloning and expression of Tobacco Etch Virus protease as a degron fusion protein
TEV protease, or rather the 27 kDa C-terminal domain of full length TEV protease con-ferring cleavage activity, has widely been used in vitro and in vivo for protein cleavage at its conserved recognition site (E-N-L-Y-F-Q-(G/S)) (reviewed in Waugh 2011). Due to its high specificity it represents a protein of a certain biotechnological importance. Transgenic T2 lines expressing Pro35S:K2:TEV were screened for the presence of the fusion protein by Western Blot analysis. For every line samples from plants grown at either 14°C or 28°C were analyzed to identify lines accumulating K2:TEV in a temperature dependent manner.
Three lines were initially identified, however, one line showed silencing effects in generation T3, leaving two lines that reproducibly showed a temperature dependent accumulation of the K2:TEV fusion protein (fig. 3.3A).
3.1.3 Cloning and expression of the BASTA resistance protein PAT as a degron fusion protein
The aim was to generate a conditionally BASTA-resistant3 A. thalianaline. Transgenic T1
were directly selected through BASTA spraying at 14°C to recover only K2:PAT expressing lines that showed the desired stability phenotype of the K2:PAT fusion protein.
K2:PAT expressing lines show neither a clear macroscopic nor protein level phenotype:. Responsive lines should survive at 14°C, but not at 28°C, due to the temper-ature dependent accumulation of the K2:PAT fusion protein. NoPro35S:K2:PAT express-ing and responsive line showexpress-ing the desired phenotype could be isolated. All lines showed resistance to BASTA regardless of growth temperature (fig. 3.3B).
To exclude possible effects of the temperature on the efficiency of the BASTA selection, also wild type plants were grown at 14°C and 28°C with or without BASTA. BASTA se-lection was not influenced by the growth temperature and killed non-transgenic plants effi-ciently under any growth condition tested (fig. S 5.4). Since it proofed impossible to isolate a responsive line expressing K2:PAT under the control of the strong ProCaMV35S, a set of new pEXPR vectors with the weaker ProUBQ10 and ProCDKA;1 driving expression was generated. These vectors were termed pAM-NOP4. As opposed to previous experiments, where only phenylalanine starting degron cassettes (K2) were used, now also version with
3BASTA (glufosinate, phosphinothricin) is a widely used herbicide. Resistance to BASTA or its active components is conferred by thebargene coding for the PAT enzyme. Two homologues have been initially isolated fromStreptomyces hygroscopicus(Thompsonet al., 1987) andStreptomyces viridochromogenes (Wohllebenet al., 1988). Since then it has been used in a wide variety of plants for research, bio-and aggrotechnological applications.
4NOP = no PAT
Figure 3.3 – Analysis of Tobacco Etch Virus (TEV) and phosphinothricin-N-acetyltransferase (PAT) as degron fusions.(A) K2:TEV expressed under the control of the strong viral ProCaMV35S results in temperature dependent accumulation of the fusion protein. CDKA;1 was used as a housekeeping gene. (B)Representative selection of different lines expressing K2:PAT. All lines showed insensitivity to BASTA independently of the growth temperature (black bar = 1mm). (C)Western blot analysis of threePro35S:K2:PAT expressing lines. Even though some changes in signal intensities and abundance could be observed overall there was no clear phenotype on protein levels which is in line with the observed insensitivity to BASTA. D Different promoters or N-termini do not alter the BASTA insensitivity phenotype (black bar = 1 mm).
Arginine (R-K2) and Leucine (L-K2) at the N-terminal were used ( pAM-NOP-ProUBQ:K2-PAT/R-K2:PAT/L-K2:PAT andpAM-NOP-ProCDKA;1:K2-PAT/R-K2:PAT/L-K2:PAT).
Selection of transgenic lines took place again at 14°C. Transgenic lines for pAm-NOP-pUBQ:K2-PAT/R-K2:PAT as well as for pAM NOP pCDKA;1:L-K2-PAT could be re-covered. These lines were screened again for a temperature responsive BASTA resistance phenotype as described previously. However, no responsive line could be isolated again (for a representative selection fig. 3.3D).
To elucidate whether the temperature insensitive resistance phenotype might still cor-relate with a responsive phenotype on protein level, meaning less of the K2:PAT fusion protein at 28 then at 14°C, selected lines were analyzed via Western Blot. Samples were harvested from three Pro35S lines (fig. 3.3B) grown at 14°C and 28°C. Multiple specific signals were identified, with the highest molecular weight one under cold growth temper-atures correlating with the expected, in silico determined, molecular weight of 46.9 kDa for the complete K2:PAT protein and with the other ones representing possible cleavage products. Interestingly, this highest band reliably disappeared under warm growth con-ditions indicating temperature dependent processing of the fusion protein (fig. 3.3C). No significant difference in overall levels of band intensities could be observed between the two growth conditions correlating with the non-responsive phenotype.
3.1.4 Addressing different N-recognins of the plant N-end rule through Phe- and Arg-starting K2:GFP constructs
Green fluorescent protein (GFP), and its various versions, are long established fluorophors, used extensively in cell biology (reviewed in Remington 2011). Pro35S:K2:GFP and Pro35S:R-K2:GFP expressing lines were screened for presence of the fusion protein at 14°C and 28°C through Western Blot analysis. Two lines expressing K2:GFP as well as two lines expressing R-K2:GFP were isolated that reliably showed the desired phenotype on protein level (data not shown). These lines were subjected to a temperature shift ex-periment (fig. 3.4A). All lines showed efficient stabilization or degradation of the fusion protein depending on the temperature shift. It is worth noting, that the R-K2:GFP ex-pressing plants performed as well as the K2:GFP exex-pressing ones indicating that the degron is also able to efficiently target R-K2:GFP for degradation through PRT6. Functionality of
the K2:GFP fusion protein was verified using a confocal laser scanning microscope. GFP fluorescence was clearly identified in the root cells of transgenic K2:GFP expressing plants.
The identity of the K2:GFP signal was confirmed through a lambda scan (fig. S 5.5).
3.1.5 The K2 cassette mediates efficient protein degradation in D.melanogaster embryonic Kc cells
To test whether the degron would also be functional in D. melanogaster, K2:GFP was expressed in embryonic Kc cells. D. melanogaster, as an example for a poikilothermic ani-mal, was chosen to demonstrate the comprehensive applicability of the degron in different organisms. 24 h post transfection cells were treated with either 1 mM of the protein trans-lation inhibitor cycloheximide (CHX) or a mock treatment and shifted for four hours to 15°C or 28°C respectively. The K2:GFP fusion protein was destabilized upon temperature shift. The effect became even more eminent when translation was disrupted through the
Figure 3.4 – K2:GFP with different termini addresses both known N-recognins in plants as well as the N-end rule in D.melanogaster. The pheno-type is a true protein one and not the result of transcriptional regulation.(A) K2:GFP adressing the Phe-as well as the Arg-branch of the N-end rule in A.thaliana. Plants were grown at 21°C temperature and shifted to 14°C or 28°C respectively. Sam-ples were taken after 0, 6, and 24 h. (B) Semi-quantitative reverse transcriptase anal-ysis shows that the degron phenotype happens on the level of stabilization rather then being a result of transcriptional regulation. (C) The degron is able to mediate degra-dation/stabilization in embryonic D.melanogaster Kc cells. (D) Functionality of the K2:GFP fusion protein in Kc cells is monitored by fluorescence microscopy.
inhibitor (fig. 3.4C). K2:GFP is also functional as indicated by microscopic analysis (fig.
3.4D).
3.1.6 Transcript analysis confirms regulation of degron levels on protein level
To ensure that the phenotypes of the different responsive degron lines is dependent on stabilization of the protein, rather then transcriptional regulation, cDNA from the K2:GFP, K2:GUS, and K2:TEV expressing lines was analyzed using a DHFR (= K2) specific primer set. Results clearly showed that the degron transcripts are not responsive to temperature, indicating a real and temperature dependent protein stabilization phenotype (fig. 3.4B).
3.1.7 Cytotoxic barnase expressed in Arabidopsis trichomes as a degron fusion is able to stir organ formation
In his master thesis, Stefan Mielke isolated a transgenic A. thaliana line expressing a K2:BAR fusion under the control of the trichome specific TRYPTICHON promoter (Pro-TRY) (Mielke, 2014). The bacterial ribonucleas (BAR) is an enzyme possessing RNase activity, synthesized and secreted by the soil bacteriumBacillus amyloliquefaciens (Buckle
& Fersht, 1994). It is potentially cytotoxic in all eukaryotic and prokaryotic cells. The ProTRY:K2:BAR expressing line showed a temperature dependent presence of trichomes.
When grown at 14°C trichomes were absent, most probably due to the cytotoxic effects of the stabilized K2:BAR fusion protein. When grown at 28°C plants appeared wild type-like (Mielke, 2014).
This line was now further characterized using different microscopic approaches to fur-ther define the K2:BAR fusion protein as a conditional cell-death module for higher plants.
A. thaliana trichomes, due to their unicellular organization and good characterization (re-viewed e.g. in Schwab et al. 2000, Hülskamp 2004), represent a versatile model system for the expression of (toxic) proteins and their effects also due to the ease of trichome observation using well established microscopic methods.
Trichome formation in the ProTRY:K2:BAR expressing line is a dynamic process controlled by temperature: Plants were grown at 14°C and 28°C respectively.
Figure 3.5 – A K2:BAR fusion protein expressed in A.thaliana trichomes controls organ formation in a temperature dependent manner.(A)Abrogation of trichomes can be stirred through conditional stability of the K2:BAR fusion protein, using a temperature stimulus. Plants were grown at either 14°C or 28°C. Cold grown plants were shifted to warm temperatures for 9 d before being shifted back to cold tem-peratures. Trichome formation on newly formed leafs was monitored (white bar = 1 cm). (B)Polarized light analysis of cleared true leafs three to six. The plants expressing ProTRY:K2:BARshow completely glabrous trichomes (black bar = 500 µm). Trichomes appear as black. (C) Agar imprints of the leaf surface of K2:BAR expressing plants.
K2:BAR expressing plants were compared with wild type plants. Plants were grown at 14°C. The trichome base of the transgenic plants has collapsed and is severely altered compared to wildtype plants. (D)DAPI staining of K2:BAR expressing plants grown at cold temperatures show a trichome phenotype similar to the gl2 phenotype. Trichome cells remain in a state of growth arrest as indicated by the still present nuclei. (E)Older leafs ofProTRY:K2:BARexpressing plants show occasional induction of trichomes sim-ilar to the previously described gl2 phenotype.
The previously reported phenotype (Mielke, 2014) was completely reproducible. Plants ex-pressingProTRY:K2:BARreliably abolished trichomes when grown at 14°C (fig. 3.5A).To test whether the trichome phenotype is dynamically responsive to temperature switches, plants grown at 14°C were switched to 28°C for nine days. New leaves clearly exhibited a normal, wild type like, trichome patterning. After a switch back to cold growth tempera-tures the K2:BAR fusion protein was stabilized and new leaves appeared glabrous5 again.
This clearly indicates that the K2:BAR fusion protein reacts dynamically to changing temperatures (fig. 3.5A).
To analyze trichome spacing and number in more detail, I used a polarized light mi-croscopy approach to visualize trichomes. Leafs of plants grown at either 14°C or 28°C were harvested and visualized. The pictures clearly confirmed the macroscopic data shown in figure 3.5A. Cold grown plants of ProTRY:K2:BAR expressing plants appear indeed glabrous (fig. 3.5B)
The K2:BAR fusion protein significantly alters the structure of the trichome base: To gain further insight into the process of trichome abolition in K2:BAR-expressing plants, agar imprints of the surface were obtained. The imprints clearly showed significant alterations in the overall structure of the trichome base. Namely, the cells around the central trichome-forming cells did not form. However, the central, barnase expressing, cell is still present in the tissue indicating a growth arrest rather then a complete ablation of the cell (fig. 3.5C).
DAPI staining of nuclei reveals a cell growth arrest similar to the gl2 mu-tant allele phenotype: To finally distinguish whether the trichome forming cells of the K2:BAR expressing plants grown at cold temperatures have collapsed or are only in a state of growth arrest I performed microscopy analysis of DAPI stained leafs (fig. 3.5D). If the cells are indeed dead they should no longer contain DNA. The analysis of stained leafs revealed that the glabrous leaf phenotype elicited through K2:BAR strongly resembles the
5glabrous = smooth, without hairs
phenotype of the GLABRA2 knockout allele (gl2) (Rerie et al. 1994, reviewed in Hül-skamp 2004). Similar to this phenotype, the cells arrested in a state of enlargement prior to forming the actual trichome indicating that the initiation and differentiation process was interrupted. Comparable to the gl2 phenotype, also some cells, even though very rarely, were able to start trichome initiation, something that was also reported for the gl2 phenotype. However, this only happened solemnly and predominantly on older leaves (fig.
3.5D).
Figure 3.6 – The K2:BAR fusion protein is degraded by the E3 ubiquitin ligase PRT1 and the phenotype is organ specific.(A)Crossing ofProTRY:K2:BAR into theprt1-1 mutant background results in a temperature independent stabilization of the fusion protein (white bar = 1 cm). (B) K2:BAR is also active in flowers (white bar = 500µm). (C) K2:BAR does not lead to increased seed abrogation under cold temperatures. Late stage siliques of wild type and ProTRY:K2:BAR expressing plants were opened and analyzed using a stereo microscope. There is no obvious difference between the control and the transgenic plants (black bar = 1 mm). (D)Transgenic and wild type seeds show no difference in mucilage deposition as made visible by a white aureole around seeds after submergence in diluted ink (black bar = 500 µm).
Genetic evidence suggests degradation of K2:BAR by the E3 ligase PRT1: To test whether the degron is indeed predominantly degraded by the E3 ligase PRT1 and to asses whether barnase activity itself might also be influenced by the changing temperature, I crossed the K2:BAR-expressing line into the PRT1 mutant background (prt1-1, Bachmair et al. 1993, Potuschak et al. 1998). Crossing into prt1-1 resulted in a temperature independent stabilization of the K2:BAR fusion protein, thus indicating that indeed the degradation by PRT1, and not potential temperature sensitivity of barnase, led to the temperature dependent toxicity phenotype (fig. 3.6A)
ProTRY is also active in flowers but not in seeds: Transcript data of the TRYP-TICHON protein obtained from the eFP browser (Winter et al. , 2007) (see also figure 5.6) shows that ProTRY is almost ubiquitiously active throughout the whole plant with expression maximals in trichomes, early flower stages as well as early seed / silique devel-opmental stages. This led to the question whether the toxic protein K2:BAR would also induce cell death in other tissues then in the leafs/trichomes. The temperature dependent expression of the K2:BAR fusion protein led to abolishment of trichomes on the flowers (fig. 3.6B). Since the expression data also hints towards expression in early seedlings and siliques the question arose whether K2:BAR expressing plants would be sterile under cold growth conditions. This was not the case.
To determine whether expression and activity of the K2:BAR fusion protein at 14°C would result in in abolished seeds, late stage siliques were opened and seeds analyzed using a standard stereo microscope. The K2:BAR expressing line did not show more abolished seeds then the Col-0 wildtype plants grown at the same temperature (fig. 3.6C).
Additionally, the seeds showed an indistinguishable mucilage deposition on the seed coat when compared to wild type seeds. Disturbed mucilage deposition in the seed coat is a reported phenotype of gl2 (Rerie et al. , 1994), indicating that the similarity to gl2 does not extend to the seed but remains exclusive for trichomes (fig. 3.6D).